Abstract
BACKGROUND: Odour baited resting boxes are simple, reliable and important tools for sampling malaria vector mosquitoes in surveillance and control programmes in different parts of Africa. To optimize the use of cow urine baited resting boxes for sampling An. arabiensis, a community-based study was conducted in Mabogini hamlet in the Lower Moshi irrigation scheme area. METHOD: Experimental designs using 3 by 3 Latin square were conducted for twenty days to evaluate the following: i) the effect of different parameters in the sampling of mosquitoes using odour baited resting boxes; ii) the performance of odour baited traps under indoor and outdoor conditions and the effect of people sleeping indoors on mosquito density; iii) the effect of position in the placement of traps on collection of mosquitoes; and, iv) the efficiency of the trap outdoors at three different distances from the house wall. One extra house served as the sentinel house to monitor species abundance using a CDC-miniature light trap. RESULTS: 8581 mosquitoes were sampled by odour baited resting boxes of which, 8051 (93.82%) were An. arabiensis and 530 (6.18%) Cx. quinquefasciatus. The light trap collected 12,420 mosquitoes, of which 9442 (76.02%) were An. arabiensis, 126 (1.01%) An. funestus group, 230 (1.85%) An. rufipes and 2622 (21.11%) Cx. quinquefasciatus. The best height for outdoor mosquitoes sampling was 15 cm and 220 cm while indoors was 105 cm. The difference in mosquito collection between different outdoor and indoor heights was statistically significant (p < 0.0001). The optimal outdoor location of odour baited resting boxes from the wall of the house was 3 m. CONCLUSIONS: The results of these studies demonstrate an optimal method for sampling during surveillance and control programmes in rural villages of highlands and arid areas of Africa using inexpensive baits and boxes.